S1 Domain RNA-Binding Protein CvfD Is a New Posttranscriptional Regulator That Mediates Cold Sensitivity, Phosphate Transport, and Virulence in Streptococcus pneumoniae D39

Abstract

Posttranscriptional gene regulation often involves RNA-binding proteins that modulate mRNA translation and/or stability either directly through protein-RNA interactions or indirectly by facilitating the annealing of small regulatory RNAs (sRNAs). The human pathogen Streptococcus pneumoniae D39 (pneumococcus) does not encode homologs to RNA-binding proteins known to be involved in promoting sRNA stability and function, such as Hfq or ProQ, even though it contains genes for at least 112 sRNAs. However, the pneumococcal genome contains genes for other RNA-binding proteins, including at least six S1 domain proteins: ribosomal protein S1 (rpsA), polynucleotide phosphorylase (pnpA), RNase R (rnr), and three proteins with unknown functions. Here, we characterize the function of one of these conserved, yet uncharacterized, S1 domain proteins, SPD_1366, which we have renamed CvfD (conserved virulence factor D), since loss of the protein results in attenuation of virulence in a murine pneumonia model. We report that deletion of cvfD impacts the expression of 144 transcripts, including the pst1 operon, encoding phosphate transport system 1 in S. pneumoniae We further show that CvfD posttranscriptionally regulates the PhoU2 master regulator of the pneumococcal dual-phosphate transport system by binding phoU2 mRNA and impacting PhoU2 translation. CvfD not only controls expression of phosphate transporter genes but also functions as a pleiotropic regulator that impacts cold sensitivity and the expression of sRNAs and genes involved in diverse cellular functions, including manganese uptake and zinc efflux. Together, our data show that CvfD exerts a broad impact on pneumococcal physiology and virulence, partly by posttranscriptional gene regulation.IMPORTANCE Recent advances have led to the identification of numerous sRNAs in the major human respiratory pathogen S. pneumoniae However, little is known about the functions of most sRNAs or RNA-binding proteins involved in RNA biology in pneumococcus. In this paper, we characterize the phenotypes and one target of the S1 domain RNA-binding protein CvfD, a homolog of general stress protein 13 identified, but not extensively characterized, in other Firmicutes species. Pneumococcal CvfD is a broadly pleiotropic regulator, whose absence results in misregulation of divalent cation homeostasis, reduced translation of the PhoU2 master regulator of phosphate uptake, altered metabolism and sRNA amounts, cold sensitivity, and attenuation of virulence. These findings underscore the critical roles of RNA biology in pneumococcal physiology and virulence.

Keywords: RNA-binding protein; cold sensitivity; phosphate uptake; pleiotropic regulation; pneumococcus; posttranscriptional regulation; virulence attenuation.

FIG 1

Multiple-sequence alignment of CvfD homologs from several Gram-positive bacteria. Alignment of the CvfD homologs from S. pneumoniae (sequence ID, ABJ54781.1), B. subtilis (sequence ID, AGG62546.1), S. epidermidis (sequence ID, OXE79056.1), S. aureus (sequence ID, KMR00465.1), Streptococcus pyogenes (sequence ID, SQG97127.1), and L. monocytogenes (sequence ID, NP_465892.1) was obtained by using ClustalOmega. The residues marked by asterisks represent conserved RNA-binding residues in the S1 domain, as predicted by Bycroft et al. (58). Residues in the S1 domain that are conserved among all of the different Gram-positive bacteria listed here are shaded in red. Frames and colors were generated using the ESPript 3.0 Web server. Columns framed in blue, amino acids with >70% similarity based on physicochemical properties; red characters, conserved amino acids; white characters, strictly conserved amino acids.

FIG 2

Growth and virulence phenotypes of the ΔcvfD mutant. (A and B) Growth characteristics of the encapsulated D39 parent strain (IU1781) and its derived ΔcvfD (IU4772) and ΔcvfD bgaA::P_ftsA cvfD⁺ (IU5508) mutant strains at 37°C and 32°C in aged (24-day-old) BHI broth. The accelerated growth rate of the ΔcvfD (IU4772) mutant at 32°C after 8 h of incubation was due to accumulation of spontaneous suppressor mutations. At least five independent growth curves were performed at 37°C or 32°C in aged BHI broth with similar results, and representative curves are shown. Average growth rates and growth yields are listed in Table S3. (C) qRT-PCR analysis was performed to determine cvfD transcript steady-state levels in the parent strain (IU1781) relative to gyrA (internal control) at 37°C and 32°C. The primers used for qRT-PCR analyses are listed in Table S5. (D and E) Western blot analysis to determine CvfD protein levels at 37°C and 32°C in an isogenic strain expressing CvfD–L-FLAG³ (IU8717). (C to E) All the strains were grown to early exponential phase in aged BHI broth, and the data points and error bars represent the means and SEM of the results of at least three independent experiments. *, P < 0.05; ns, not significantly different. (F) Survival curve analysis showing disease progression in an invasive model of pneumonia. ICR male mice were inoculated intranasally with ∼10⁷ CFU in 50 μl inoculum of either the D39 parent strain (IU1781) or its derived isogenic mutants (IU4772 [ΔcvfD] or IU5508 [ΔcvfD bgaA::P_ftsA cvfD⁺; complemented strain]). Eight animals were infected per strain, and disease progression was followed in real time by a survival curve analysis (see Materials and Methods). The survival curves were analyzed by Kaplan-Meier statistics and log rank tests to determine P values.

FIG 3

CvfD posttranscriptionally regulates the PhoU2 master regulator. (A) Volcano plot showing genome-wide changes in transcript levels from ORFs in a ΔcvfD mutant relative to the D39 parent strain. RNA was extracted from exponential-growth-phase cultures of the wild-type D39 parent (IU3116; D39 rpsL1 CEP::P_c-[Kan^r-rpsL⁺]) and its derived isogenic ΔcvfD mutant (IU5506; D39 ΔcvfD rpsL1 CEP::P_c-[Kan^r-rpsL⁺]) in triplicate and prepared for mRNA-seq analysis as described in Materials and Methods. The red and blue dots represent genes with relative transcript changes over a 1.8-fold cutoff (log₂ change = 0.85-fold) and a 3-fold cutoff (log₂ change =1.5-fold), respectively, with an adjusted P value cutoff of 0.05, which were considered to be significant changes. The genes displaying changes below the above-mentioned values and P value cutoff were considered to have nonsignificant changes and are represented by black dots. The x axis shows gene fold changes, and the y axis shows the corresponding P values plotted on a logarithmic scale. Genes that were significantly upregulated or downregulated in a ΔcvfD mutant compared to the parent are listed in Table 1. (B, C, and D) qRT-PCR analysis was used to determine pstS1, phoU1, and phoU2 transcript levels in a WT D39 parent (IU1781) and its derived isogenic mutants (IU4772 [ΔcvfD] and IU5508 [ΔcvfD bgaA::P_ftsA cvfD⁺]). Transcript signal intensities were normalized to the gyrA transcript level, which served as the internal control, and subsequently, the expression level relative to the WT strain, IU1781, which was set to 1, was calculated. (E) Western blot analysis to determine PhoU2 protein levels. Samples were prepared for Western blotting from exponential-phase cultures of a wild-type strain (IU1781) and its derived strains IU8675 (phoU2-HA), IU8722 (phoU2-HA ΔcvfD), and IU8719 (phoU2-HA ΔcvfD bgaA::P_ftsA cvfD⁺), with anti-HA antibody. Representative Western blots are shown. (F) Quantification of PhoU2 levels from the Western blots shown in panel E. PhoU2 expression levels were calculated relative to that in a WT strain, IU1781, which was set to 1. (G) RIP–qRT-PCR was used to determine interactions between phoU2 mRNA and CvfD. The strains used in the RIP experiments were derived from an unencapsulated isogenic variant of D39. CvfD was immunoprecipitated from exponential-phase cultures of the wild type (IU1945) and an isogenic strain expressing CvfD–L-FLAG³ from its native locus (IU5809) using anti-FLAG antibody (see Materials and Methods). RNAs extracted from the elution fractions were reverse transcribed to cDNA and subsequently used as templates for qRT-PCR to determine the relative levels of the phoU2 transcript coimmunoprecipitated from lysates of a strain expressing CvfD–L-FLAG³ compared to that of an untagged strain (WT); the phoU2 expression level was set to 1, and 16S rRNA was used as an internal control. (B, C, D, F, and G) The data represent the means and SEM of the results of at least three independent experiments. *, P < 0.05; ***, P < 0.0001; ns, not significantly different. The primers used for qRT-PCR analyses are listed in Table S5.

FIG 4

Loss of cvfD perturbs metal homeostasis in S. pneumoniae D39. (A and B) qRT-PCR analysis was used to determine psaA and czcD transcript levels in a wild-type D39 parent (IU1781) and its derived isogenic mutants (IU4772 [ΔcvfD] and IU5508 [ΔcvfD bgaA::P_ftsA cvfD⁺]) as described in the legend to Fig. 3. The data and error bars represent the means and SEM of the results of at least three independent experiments. *, P < 0.05; **, P < 0.01; ns, not significantly different. (C) Growth characteristics of a ΔcvfD mutant (IU4772) in BHI broth with no MnSO₄ or ZnCl₂ addition, addition of 0.5 mM MnSO₄ (+ Mn), or addition of 0.5 mM MnSO₄ and 0.2 mM ZnCl₂ (+ Mn + Zn) relative to the wild-type parent (IU1781) at 37°C. At least four independent growth curves were performed under the above-mentioned conditions, except for the condition where both MnSO₄ and ZnCl₂ were added (+ Mn + Zn), for which two independent experiments were performed, and representative curves are shown. (D) Growth characteristics of a ΔcvfD mutant (IU4772) in BHI broth in the presence or absence of 0.2 mM ZnCl₂ relative to the wild-type parent (IU1781) at 37°C. Representative curves from the results of at least six independent growth experiments are shown. (C and D) Average growth rates and growth yields are listed in Table S3.

FIG 5

PhoU2-HA protein is more stable in a ΔcvfD strain than in a cvfD⁺ strain. (A) Representative growth and Cm response curves of the encapsulated D39 parent strain (IU1781; WT) and strains IU8675 (phoU2-HA) and IU8722 (phoU2-HA ΔcvfD). The arrows indicate addition of Cm at an OD₆₂₀ of ∼0.1. (B) Semilogarithmic decay plot of relative PhoU2-HA protein amounts versus time after Cm addition obtained from the results of three independent experiments. The half-lives of PhoU2-HA proteins in IU8675 (phoU2-HA) and IU8722 (phoU2-HA ΔcvfD) were determined after treatment with Cm, a protein translation inhibitor, as described in Materials and Methods. Each data point represents the mean ± SEM (where error bars are not visible, they are smaller than the symbol) of PhoU2-HA protein at each time point after Cm treatment. Half-life values, 95% confidence intervals of the half-life values, and R² values of the decay curves were obtained from nonlinear-regression analysis with GraphPad Prism as described in Materials and Methods. The amount of PhoU2-HA protein in the phoU2-HA ΔcvfD strain at T₀ relative to that in the phoU2-HA cvfD⁺ strain at T₀ was determined to be 0.48 ± 0.03 (mean ± SEM) from the results of three independent experiments (see Fig. S4).

FIG 6

Loss of capsule suppresses the observed growth defect of a ΔcvfD mutant. Shown are the growth characteristics of the unencapsulated D39 cps2E(ΔA) strain (IU3309), the encapsulated D39 cps⁺ cvfD⁺ parent strain (IU1781), and their derived ΔcvfD cps2E(ΔA) (IU8396) and cps⁺ ΔcvfD (IU4772) mutant strains at 37°C (A) and at 32°C (B). At least two independent growth curves were performed at 37°C or 32°C, and representative curves are shown. Average growth rates and growth yields are listed in Table S3.

FIG 7

Overview of the regulatory scope of CvfD in S. pneumoniae. Major pathways that are regulated by CvfD are indicated.